✨ Set default command for each dataset

README.md

@@ -50,11 +50,9 @@ seq2rel-ds --help
 To preprocess a dataset (and in most cases, download it), call one of the commands, e.g.
 
 ```bash
-seq2rel-ds preprocess cdr main "path/to/cdr"
+seq2rel-ds cdr "path/to/cdr"
 ```
 
-> Note, you have to include `main` because [`typer`](https://typer.tiangolo.com/) does not support default commands.
-
 This will create the preprocessed `tsv` files under the specified output directory, e.g.
 
 ```

seq2rel_ds/cdr.py

@@ -124,7 +124,7 @@ def _preprocess(
     return seq2rel_annotations
 
 
-@app.command()
+@app.callback(invoke_without_command=True)
 def main(
     output_dir: Path = typer.Argument(..., help="Directory path to save the preprocessed data."),
     sort_rels: bool = typer.Option(

seq2rel_ds/dgm.py

@@ -87,7 +87,7 @@ def _preprocess(
     return seq2rel_annotations
 
 
-@app.command()
+@app.callback(invoke_without_command=True)
 def main(
     input_dir: Path = typer.Argument(..., help="Path to a local copy of the DGM corpus."),
     output_dir: Path = typer.Argument(..., help="Directory path to save the preprocessed data."),
@@ -111,6 +111,7 @@ def main(
     ),
 ) -> None:
     """Download and preprocess the DGM corpus for use with seq2rel.
+
     The corpus can be downloaded at: https://hanover.azurewebsites.net/downloads/naacl2019.aspx.
     Provide this path as argument `input_dir` to this command. More details about the corpus can be
     found here: https://arxiv.org/abs/1904.02347. Note that we use the paragraph-length text.

seq2rel_ds/docred.py

@@ -83,14 +83,18 @@ def _preprocess(
     return seq2rel_annotations
 
 
-@app.command()
+@app.callback(invoke_without_command=True)
 def main(
     output_dir: Path = typer.Argument(..., help="Directory path to save the preprocessed data."),
     sort_rels: bool = typer.Option(
         True, help="Sort relations according to order of first appearance."
     ),
 ) -> None:
-    """Download and preprocess the DocRED corpus for use with seq2rel."""
+    """Download and preprocess the DocRED corpus for use with seq2rel.
+
+    This is the end-to-end split provided by https://arxiv.org/abs/2102.05980, which can be
+    accessed here: http://lavis.cs.hs-rm.de/storage/jerex/public/datasets/docred_joint/.
+    """
     msg.divider("Preprocessing DocRED")
 
     with msg.loading("Downloading corpus..."):

seq2rel_ds/gda.py

@@ -106,7 +106,7 @@ def _preprocess(
     return seq2rel_annotations
 
 
-@app.command()
+@app.callback(invoke_without_command=True)
 def main(
     output_dir: Path = typer.Argument(..., help="Directory path to save the preprocessed data."),
     sort_rels: bool = typer.Option(